We study whether patient-side cost-sharing prompts patients to consume healthcare more efficiently by reducing spending on low-value services. We find some evidence that offering employees high-deductible plans disproportionately reduces low-value spending relative to overall spending, indicating that deductibles may be a way to incentivize value-based decision making.

We study whether patient-side cost-sharing prompts patients to consume healthcare more efficiently by reducing spending on low-value services. We find that consumer-directed health plans reduce spending on low-value services, though these reductions are proportional to overall spending reductions.

A comparison of intensity of post-acute care utilization following hospitalization between Medicare Advantage and Traditional Medicare beneficiaries. Despite having similar or improved health outcomes, MA patients spent less time in post-acute care settings.

The COVID-19 pandemic resulted in an explosion in demand for telemedicine and with it myriad new, untested telemedicine policies. The effects of these policies are largely unknown, and the shift to telemedicine can have longstanding impacts. We isolate the impact of a policy which reduced cost-sharing for telemedicine services from April-September 2020, estimating changes in telemedicine utilization, the elasticity of demand, and the degree of substitution with in-person services. Additionally, we study the effects of shifting to the virtual environment on mental health treatment decisions, adverse health events, and overall spending. We estimate the elasticity of telemedicine to be -0.21, and that it is used as a near perfect substitute for in-person care, resulting in no net increase in utilization. For those with mental illness, we find that increasing telemedicine use results in more psychotherapy sessions and a reduction in the use of prescription medications. Finally, we show that higher rates of telemedicine use lead to a modest reduction in spending and a higher probability of receiving emergency department care

There is much debate about the effects of pharmaceutical direct to consumer advertising (DTCA) on health care use. In this paper, we inform this debate by examining the effects of DTCA on office visits, as well as treatment courses resulting from those visits, for five common chronic conditions (hypertension, hyperlipidemia, diabetes, depression, and osteoporosis). In particular, we examine whether office visits result in use of drug therapy and/or continued office visits over time. We test these questions by combining data on pharmaceutical advertising from Nielsen with claims data from 40 large national employers, covering 18 million person-years (2004-2010). We analyze a non-elderly population by exploiting plausibly exogenous variation in advertising exposure across areas by exploiting the fact that the implementation of Medicare prescription drug coverage led to larger increases in advertising in areas with high elderly share of population compared to low elderly share areas. We find that advertising increases the number of office visits for the non-elderly for the advertised condition. We also find substantial spillovers – a large share of the increased office visits from advertising do not result in use of drugs or are associated with use of generic drugs. Finally, we find that the increase in office visits due to DTCA is associated with multiple consecutive follow up visits over time.